Retarded persons are known to be deficient in a variety of learning and memory tasks where strategic processing is necessary for optimal performance. The present research will study the training of different strategies in recall tasks with the initial focus on semantic coding and comparisons between semantic encoding and cumulative rehearsal. Exploratory experiments will investigate what type of semantic encoding best facilitates learning and retention in both training and transfer. Once the "best" semantic encoding strategy is selected it will be compared with cumulative rehearsal in terms of relative effectiveness in facilitating recall. In addition to recall being a major dependent variable of interest, efficiency of strategy training e.g. average study time/number of items recalled, will also be analyzed. A primary objective will be to determine an optimally efficient and effective strategy for a given task with subjects of particular characteristics. Thus, individual differences will be of interest. Later studies will investigate flexibility in strategy selection and training of the executive function. The final aim is to extend strategy training to classroom situations with tasks of academic relevance.